Diffusion Acceleration with Gaussian process Estimated Reconstruction (DAGER)

Abstract

Total word count = Approx. 7500Running title: Diffusion Acceleration with DAGER Abstract Purpose: Image acceleration provides multiple benefits to diffusion MRI (dMRI), with in-plane acceleration reducing distortion and slice-wise acceleration increasing the number of directions that can be acquired in a given scan time. However, as acceleration factors increase, the reconstruction problem becomes ill-conditioned, particularly when using both in-plane acceleration and simultaneous multi-slice (SMS) imaging. In this work, we develop a novel reconstruction method for in-vivo MRI acquisition that provides acceleration beyond what conventional techniques can achieve. Theory and Methods:We propose to constrain the reconstruction in the spatial (k) domain by incorporating information from the angular (q) domain. This approach exploits smoothness of the signal in q-space using Gaussian processes, as has previously been exploited in postreconstruction analysis. We demonstrate in-plane acceleration exceeding the theoretical parallel imaging limits, and SMS combined with in-plane acceleration at a total factor of 12. This reconstruction is cast within a Bayesian framework that incorporates estimation of smoothness hyper-parameters, with no need for manual tuning.Results: Simulations and in vivo results demonstrate superior performance of the proposed method compared with conventional parallel imaging methods. These improvements are achieved without loss of spatial or angular resolution and require only a minor modification to standard pulse sequences. Conclusion:The proposed method provides improvements over existing methods for diffusion acceleration, particularly for high SMS acceleration with in-plane undersampling.

High‐fidelity, accelerated whole‐brain submillimeter in vivo diffusion MRI using gSlider‐spherical ridgelets (gSlider‐SR)

“…Although not a direct comparison on the same datasets, angular errors using gSlider‐SR are consistently lower than those reported in, and comparable to those obtained with simulations (2X).…”

High-fidelity, high-spatial-resolution diffusion MRI of the ex-vivo whole human brain on the 3T Connectom scanner using structured low-rank EPI ghost correction

“…The fact that ghosting varies across diffusion direction suggests that it may be advisable to reconstruct the whole set of dMRI images simultaneously by exploiting similarities of kspace acquired at different diffusion directions or q-space values. Those approaches are becoming increasingly popular in the dMRI literature MRI (Ramos-Llordén, et al, 2020) (Wu, Koopmans, Andersson, & Miller, 2019) (Mani, Magnotta, & Jacob, 2021) (Hu, et al, 2020) (Ramos-Llordén, Aja-Fernández, Liao, Setsompop, & Rathi, 2019) (Haldar, et al, 2013) (Wu, Koopmans, Andersson, & Miller, 2019) (Wu, et al, 2014). Exploratory analyses along these lines are an important aspect of our future work.…”

Diffusion‐PEPTIDE: Distortion‐ and blurring‐free diffusion imaging with self‐navigated motion‐correction and relaxometry capabilities

“…Simultaneous multi‐slice 44‐46 or gSlider simultaneous multi‐slab 47,48 acquisitions could also be incorporated with PEPTIDE to provide improvement in acquisition speed and SNR efficiency. Other techniques could also be incorporated to aid in reduction of the diffusion acquisition time itself 49,50 . Adding an inversion‐preparation scheme would provide T 1 relaxometry information for more complete signal characterization, which could be used to perform synthetic imaging of different contrasts 51,52 or for higher dimensionality joint diffusion‐relaxometry analysis.…”